huggingface/transformers

Pull Requests Commits

Remove train_fsdp_tp_torchtitan_style.py

3outeille committed 57 days ago

34db8405

Merge branch 'distributed_api' of https://github.com/huggingface/transformers into distributed_api

3outeille committed 57 days ago

e7832312

Add distributed training scripts

3outeille committed 57 days ago

c33873e7

Merge branch 'main' into distributed_api

3outeille committed 57 days ago

Verified eeefc9e8

Fix #45305 + add regression test GAS (#45349)

florian6973 committed 57 days ago

Verified 20b6f538

Update `trackio` integration to use Buckets and "freeze" Space after training (#45329)

abidlabs committed 57 days ago

Verified 7afee868

Merge remote-tracking branch 'origin/main' into distributed_api

3outeille committed 57 days ago

45a01a54

fix(qwen3_moe): correct return type annotation on Qwen3MoeSparseMoeBlock.forward (#45352)

RudrenduPaul committed 57 days ago

Verified 357f414c

Fix: NotebookProgressCallback crash when evaluating with the Trainer (#44949)

Charly21r committed 57 days ago

Verified 0b5dbfc9

Merge branch 'main' into distributed_api

3outeille committed 57 days ago

Verified b7ec958e

docs: fix 5 docstring errors in Gemma3nTextConfig (typos, grammar, formatting) (#45370)

RudrenduPaul committed 57 days ago

Verified 6db69b91

Less unnecessary RoPE warnings (#45289)

zucchini-nlp committed 57 days ago

Verified 19a7c100

[`Tokenizers`] Move gpt sw3 tokenizer out (#45404)

vasqu committed 57 days ago

Verified 2c498dfd

Fix unintended Hub metadata calls from _patch_mistral_regex (#43603)

vaibhav-research committed 58 days ago

Verified def8e6a2

Fix MoE routers returning probabilities instead of logits (#45131)

yacinemebarki committed 58 days ago

Verified d2941693

Fix NaN weights on non-rank-0 FSDP processes (#45050)

albertvillanova committed 58 days ago

Verified ff49f7c4

remove cache file from tree (#45392)

tarekziade committed 58 days ago

Verified a001f344

[docs] training on specific hardware (#44799)

stevhliu committed 60 days ago

Verified a5533957

[docs] zero + sequence parallelism (#44605)

stevhliu committed 60 days ago

Verified 09d91db3

Fix vlm weight mappings (#45358)

Cyrilvallez committed 60 days ago

Verified f3a68c4c

Copy the template resolution logic from the base apply_chat_template to Voxtral (#45117)

Rocketknight1 committed 60 days ago

Verified 5c7190f0

add kwargs to all methods in the CallbackHandler class (#45353)

wilnn committed 60 days ago

Verified 520005e3

Close file handler (#45187)

ydshieh committed 60 days ago

Verified c9f1b95a

fix: restore mypy type checking for PreTrainedConfig subclasses (#45071) (#45240)

shhKnight30 committed 61 days ago

Verified ce83f16d

`cohere_asr`: fix device issue for `test_model_parallel_beam_search` (#45214)

kaixuanliu committed 61 days ago

Verified e07692b3

Fix AttributeError in Gemma3ForConditionalGeneration and Gemma3ForSequenceClassification when config.return_dict=False (#45277)

kamalrajkannan78 committed 61 days ago

Verified 507a37d1

fix bug for videomt model device mismatch (#45204)

kaixuanliu committed 61 days ago

Verified 4fd862fc

fix gemma4 gradient accumulation loss and last token incorrect labels (#45354)

winglian committed 61 days ago

Verified 47d77657

Logger has `[transformers]` prefix in non-verbose mode (#45316)

zucchini-nlp committed 61 days ago

Verified d13b44b6

Fix AttributeError in AssistantToTargetTranslator.unmap_input_ids with cross-vocab models (#45320)

Regata3010 committed 61 days ago

Verified c43f15c7

Older